The development of improved pharmacotherapies for the treatment of cocaine and polydrug abuse with concomitant reduction of needle-sharing and risk for AIDS is of major public health importance. We discovered that buprenorphine, an opioid mixed agonist-antagonist, selectively reduced cocaine self-administration by rhesus monkeys. Ongoing clinical trials indicate that buprenorphine significantly reduces cocaine abuse, heroin abuse and needle-sharing in men who have abused these drugs daily for over 10 years and meet DSM-III-R criteria for dual dependence on cocaine and opiates. Clinical confirmation of primate data indicates the value of this pre-clinical model for studying the mechanisms underlying buprenorphine's effects on cocaine self-administration. We propose to continue studies of the behavioral pharmacology of buprenorphine and clarify its interactions with cocaine. Operant behavioral procedures will be used to maintain cocaine and food self-- administration in this established primate model of drug self-administration. The relative contribution of the agonist and the antagonist components of this opioid mixed agonist-antagonist to its effects on the cocaine self-administration dose-response curve will be examined by comparing the effects of very low (agonist range) and high (antagonist range) doses of buprenorphine. We also propose to study the effects of acute and chronic buprenorphine administration in combination with cocaine on food-maintained responding to evaluate their interactive effects on a non-drug reinforcer. Although buprenorphine is classified as a partial mu agonist, it also has activity at other putative opioid receptor sites and acts as a kappa antagonist as well as a mu agonist and antagonist. We propose to examine the effects of prototypic mu and kappa opioids alone and in combination with buprenorphine on the cocaine reinforcing dose-response curve. Data obtained should clarify the relative contribution of mu and kappa opioid activity to the behavioral interactions between buprenorphine and cocaine. The biological concomitants of buprenorphine-cocaine interactions may involve the hypothalamic-pituitary adrenal axis. Buprenorphine significantly reduced cocaine's stimulation of ACTH and reports of euphoria in man. We propose to study cocaine's effects on pulsatile release of ACTH in monkey and evaluate the effects of acute and chronic buprenorphine pre-treatment on cocaine-stimulated ACTH and cortisol release. These studies should contribute to the understanding of the biological and behavioral bases of buprenorphine-cocaine interactions and suggest new directions for development of improved pharmacotherapies.